On the Reliability of User-Centric Evaluation of Conversational Recommender Systems

• URL: http://arxiv.org/abs/2602.17264v1
• Date: Thu, 19 Feb 2026 11:10:11 GMT
• Title: On the Reliability of User-Centric Evaluation of Conversational Recommender Systems
• Authors: Michael Müller, Amir Reza Mohammadi, Andreas Peintner, Beatriz Barroso Gstrein, Günther Specht, Eva Zangerle,
• Abstract summary: We present a large-scale empirical study investigating the reliability of user-centric CRS evaluation on static dialogue transcripts.<n>We collect 1,053 annotations from 124 crowd workers on 200 ReDial dialogues using the 18-dimensional CRS-Que framework.<n>Our results show that utilitarian and outcome-oriented dimensions such as accuracy, usefulness, and satisfaction achieve moderate reliability under aggregation.<n>Many dimensions collapse into a single global quality signal, revealing a strong halo effect in third-party judgments.
• Score: 0.9112926574395824
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: User-centric evaluation has become a key paradigm for assessing Conversational Recommender Systems (CRS), aiming to capture subjective qualities such as satisfaction, trust, and rapport. To enable scalable evaluation, recent work increasingly relies on third-party annotations of static dialogue logs by crowd workers or large language models. However, the reliability of this practice remains largely unexamined. In this paper, we present a large-scale empirical study investigating the reliability and structure of user-centric CRS evaluation on static dialogue transcripts. We collected 1,053 annotations from 124 crowd workers on 200 ReDial dialogues using the 18-dimensional CRS-Que framework. Using random-effects reliability models and correlation analysis, we quantify the stability of individual dimensions and their interdependencies. Our results show that utilitarian and outcome-oriented dimensions such as accuracy, usefulness, and satisfaction achieve moderate reliability under aggregation, whereas socially grounded constructs such as humanness and rapport are substantially less reliable. Furthermore, many dimensions collapse into a single global quality signal, revealing a strong halo effect in third-party judgments. These findings challenge the validity of single-annotator and LLM-based evaluation protocols and motivate the need for multi-rater aggregation and dimension reduction in offline CRS evaluation.

Related papers

• Confidence Estimation for LLMs in Multi-turn Interactions [48.081802290688394]
  This work presents the first systematic study of confidence estimation in multi-turn interactions.<n>We establish a formal evaluation framework grounded in two key desideratas: per-turn calibration and monotonicity of confidence.<n>Our work provides a foundational methodology for developing more reliable and trustworthy conversational agents.
  arXiv  Detail & Related papers  (2026-01-05T14:58:04Z)
• When Can We Trust LLMs in Mental Health? Large-Scale Benchmarks for Reliable LLM Evaluation [14.24379104658635]
  MentalBench-100k consolidates 10,000 one-turn conversations from three real scenarios datasets, each paired with nine LLM-generated responses.<n>MentalBench-70kreframes evaluation by comparing four high-performing LLM judges with human experts across 70,000 ratings on seven attributes.<n>Our analysis reveals systematic inflation by LLM judges, strong reliability for cognitive attributes such as guidance and informativeness, reduced precision for empathy, and some unreliability in safety and relevance.
  arXiv  Detail & Related papers  (2025-10-21T19:21:21Z)
• Limitations of Current Evaluation Practices for Conversational Recommender Systems and the Potential of User Simulation [19.14733504795247]
  This paper critically examines current evaluation practices for conversational recommender systems (CRSs)<n>We identify two key limitations: the over-reliance on static test collections and the inadequacy of existing evaluation metrics.<n>We propose novel evaluation metrics, based on a general reward/cost framework, designed to better align with real user satisfaction.
  arXiv  Detail & Related papers  (2025-10-07T07:12:47Z)
• TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them [58.04324690859212]
  Large Language Models (LLMs) as automated evaluators (LLM-as-a-judge) has revealed critical inconsistencies in current evaluation frameworks.<n>We identify two fundamental types of inconsistencies: Score-Comparison Inconsistency and Pairwise Transitivity Inconsistency.<n>We propose TrustJudge, a probabilistic framework that addresses these limitations through two key innovations.
  arXiv  Detail & Related papers  (2025-09-25T13:04:29Z)
• Structured Relevance Assessment for Robust Retrieval-Augmented Language Models [0.0]
  We introduce a framework for structured relevance assessment that enhances RALM robustness.<n>Our approach employs a multi-dimensional scoring system that considers both semantic matching and source reliability.<n>Preliminary evaluations demonstrate significant reductions in hallucination rates and improved transparency in reasoning processes.
  arXiv  Detail & Related papers  (2025-07-28T19:20:04Z)
• RAG-Zeval: Towards Robust and Interpretable Evaluation on RAG Responses through End-to-End Rule-Guided Reasoning [64.46921169261852]
  RAG-Zeval is a novel end-to-end framework that formulates faithfulness and correctness evaluation as a rule-guided reasoning task.<n>Our approach trains evaluators with reinforcement learning, facilitating compact models to generate comprehensive and sound assessments.<n>Experiments demonstrate RAG-Zeval's superior performance, achieving the strongest correlation with human judgments.
  arXiv  Detail & Related papers  (2025-05-28T14:55:33Z)
• ConSiDERS-The-Human Evaluation Framework: Rethinking Human Evaluation for Generative Large Language Models [53.00812898384698]
  We argue that human evaluation of generative large language models (LLMs) should be a multidisciplinary undertaking. We highlight how cognitive biases can conflate fluent information and truthfulness, and how cognitive uncertainty affects the reliability of rating scores such as Likert. We propose the ConSiDERS-The-Human evaluation framework consisting of 6 pillars -- Consistency, Scoring Criteria, Differentiating, User Experience, Responsible, and Scalability.
  arXiv  Detail & Related papers  (2024-05-28T22:45:28Z)
• DCR-Consistency: Divide-Conquer-Reasoning for Consistency Evaluation and Improvement of Large Language Models [4.953092503184905]
  This work proposes DCR, an automated framework for evaluating and improving the consistency of Large Language Models (LLMs) generated texts. We introduce an automatic metric converter (AMC) that translates the output from DCE into an interpretable numeric score. Our approach also substantially reduces nearly 90% of output inconsistencies, showing promise for effective hallucination mitigation.
  arXiv  Detail & Related papers  (2024-01-04T08:34:16Z)
• C-PMI: Conditional Pointwise Mutual Information for Turn-level Dialogue Evaluation [68.59356746305255]
  We propose a novel model-agnostic approach to measure the turn-level interaction between the system and the user. Our approach significantly improves the correlation with human judgment compared with existing evaluation systems.
  arXiv  Detail & Related papers  (2023-06-27T06:58:03Z)
• Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation [136.16507050034755]
  Existing human evaluation studies for summarization either exhibit a low inter-annotator agreement or have insufficient scale. We propose a modified summarization salience protocol, Atomic Content Units (ACUs), which is based on fine-grained semantic units. We curate the Robust Summarization Evaluation (RoSE) benchmark, a large human evaluation dataset consisting of 22,000 summary-level annotations over 28 top-performing systems.
  arXiv  Detail & Related papers  (2022-12-15T17:26:05Z)
• Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approach [84.02388020258141]
  We propose a new framework named ENIGMA for estimating human evaluation scores based on off-policy evaluation in reinforcement learning. ENIGMA only requires a handful of pre-collected experience data, and therefore does not involve human interaction with the target policy during the evaluation. Our experiments show that ENIGMA significantly outperforms existing methods in terms of correlation with human evaluation scores.
  arXiv  Detail & Related papers  (2021-02-20T03:29:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.